Customizing Emotional Support: How Do Individuals Construct and Interact With LLM-Powered Chatbots

TL;DR

The paper addresses how individuals construct and interact with LLM-powered chatbots to obtain personalized emotional support. It uses a research-through-design approach withChatLab, a prototyped platform, and a 7–10 day field study involving 22 participants to explore persona customization, voice/avatar enrichment, and memory management. Key findings show diverse persona types (emotional support, stress coping, intellectual discourse, self-reflection, and therapy), extensive use of voice/avatar cues to shape dynamics, and design opportunities around learning signals, memory control, and community sharing. These insights inform design implications for more individualized, multimodal emotional support tools leveraging generative AI, with emphasis on user agency and reflective practice.

Abstract

Personalized support is essential to fulfill individuals' emotional needs and sustain their mental well-being. Large language models (LLMs), with great customization flexibility, hold promises to enable individuals to create their own emotional support agents. In this work, we developed ChatLab, where users could construct LLM-powered chatbots with additional interaction features including voices and avatars. Using a Research through Design approach, we conducted a week-long field study followed by interviews and design activities (N = 22), which uncovered how participants created diverse chatbot personas for emotional reliance, confronting stressors, connecting to intellectual discourse, reflecting mirrored selves, etc. We found that participants actively enriched the personas they constructed, shaping the dynamics between themselves and the chatbot to foster open and honest conversations. They also suggested other customizable features, such as integrating online activities and adjustable memory settings. Based on these findings, we discuss opportunities for enhancing personalized emotional support through emerging AI technologies.

Figures (2)

• Figure 1: The Chatting interface (C1) with an example of participant P22's customized settings (C2) and one of their Experience Diary entries (E). Note that the participant chose voice as the interaction mode; for convenience, we transcribed the audio into text. The original interface, conversations, and diary entries were in Chinese and were later translated into English.
• Figure 2: Examples of designs created by our participants on Excalidraw, with the notes translated from Chinese to English. The examples provided here were from P1 (E1, a mobile app that can upload current physical scenes as the chatting background), P21 (E2, a proactive companion that can sense their bodily and emotional states), and P13 (E3, an AI assistant guiding the customization process), respectively.